bisected trimeters in attic tragedy

7/28/2019 Bisected Trimeters in Attic Tragedy

1/23

Bisected Trimeters in Attic TragedyAuthor(s): Thomas D. GoodellSource: Classical Philology, Vol. 1, No. 2 (Apr., 1906), pp. 145-166Published by: The University of Chicago PressStable URL: http://www.jstor.org/stable/262258 .

Accessed: 13/07/2013 18:00

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .

http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to

Classical Philology.

http://www.jstor.org

This content downloaded from 189.123.99.105 on Sat, 13 Jul 2013 18:00:13 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=ucpresshttp://www.jstor.org/stable/262258?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/262258?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=ucpress


2/23

BISECTED TRIMETERS IN ATTIC TRAGEDYBR THOMAS D. GOODELL

It is part of the modern metrical tradition that iambictrimeters divided by a caesura in the middle are rather rare,and are more or less faulty. A few quotations of fairly recentdate will be sufficient as illustrations. Thus Rossbach: "Stillmore is the caesura in the middle of the verse avoided, becausethis gives the trimeter an unrhythmical division."' For Pers. 465W4en ' dvoj1w$,ev KaK(W)VoSpv /3faosand 509

PYK'qV 7rfpaaaVT0; JoXAt 7roXXw^ roTv(and Eur. Suppl. 699Kat rTVp4TaMTavWES /IEOOV 7ravTa aTXpaLov

he accepts G. Hermann's remark that the acppvO,uMawas deliber-ately chosen um den Inhalt malend hervorzuheben. A similar"excuse'" for the faulty structure frequently turns up in thenotes of school editions. Gleditsch, recognizing the existenceof verses thus divided and citing Soph. El. 1036

arqTLas juev OV, 7rpoiu77'tas & aovthen adds: Doch schwindet das Anstossige der caesura media,wenn eine Elisionssilbe fiber den dritten Fuss hinausgreifl,2 forwhich he cites Ag. 20

vvv 8' Cwrvx'XSEVOLT' 'raXXay"q rovwv.That in other cases the caesura media is objectionable is assumed.Christ is more cautious not to go beyond the facts; he merelyadmits3 the occurrence of several (mehrere) such verses, "espe-cially in Aischylos and Aristophanes." Masqueray, after statingthat the most frequent caesura of the trimeter is the penthe-mimeral, the hephthemimeral being far less common, then displaysevident reluctance to allow any other.' So in the line

(yT) 7rpo0fr?TqS ao Xoywv yCvnoouaaIGriech. Metrik3, p. 222. 2Metrik3, p. 141. 9Metrik2, p. 334.4 Traite de metrique grecque, ?? 153-63.

[CLASSICAL PHILOLOGY I, April, 1906] 145

http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/23

146 THOMASD. GOODELLhe puts the caesura between 7rpo4nT'-qqnd the enclitic; in

/?)T?)p aAyc7 T7', TpL'TOVS' 'IvO xopovhe put it between 'Awyav and o4 In the linesAAA(R) 04AX UaT7Tow, E7fret as uav0aLv(and

(XTT OVK av CV&Kws y' aTLwaLtLtOLTOOthe says we must suppose that the elision was not made. There isno apparent reason for such treatment of these lines, other thanthe supposed rule that the media caesura is to be admitted onlyunder the direst necessity. Yet he is forced to admit cases ofthat caesura, citing as examples of it

Ws Ev utaU 7rX?1y?1aTCEg0aptc 7roXvs.As regards possible explanations of the significance of such"irregular" verses Masqueray is willing to go no farther thanNous n'en savons rien. Constatons simplement les faits. Thelike assumption of rarity and of disagreeableness appears inWecklein's note on Prom. 640

OVK OL' 07Tr)s Vf LV 'rLtraat ys XpY."Verses lacking the usual caesura, and dividing themselves intotwo halves, are not frequent in Aischylos. The greatest number(seven) occurs in the Persians. In this place the ill effect issomewhat relieved by a pause after ov'ico18'." So on Pers. 254he notes that here, as in 468 and 512, "the heaviness of therhythm" is in harmony with the thought. Still more positivelyKaibel, in his edition of the Sophoklean Elektra, remarks on1036 (quoted above):

The trimeter broken in the middle is all the hasslicher in that theequal division is strongly brought out by the antithesis of thought andby the rime; much less disagreeable,for example, is Phil. 1021N , . sf S 87 , \ -,oa /AvACV OaE WVV y o aXyvvoMat

ToUT' avO' oTLboth becausethere is no rime and becausethe thought does not end withdXyvvo,uat. Yet in Phil. 1009

cVacLoV bLEVaov, KacTa'LOV' E/uOvSophokles has another verse quite like El. 1036. On the greater careshown by Euripides cf. WilamowitzEur. Her. JI2, p. 170.



4/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 147This note of Wilamowitz is worth quoting at some length; it

is on line 754:Had Euripides written 7ra'aaKa8uov y , &oXXAv,atSo'Xw,he wouldhave written, not indeed an incorrect line, but a cacophonousone (einenmislautenden vers), wie er es nie getan hat. For it sounded very dis-agreeable (hasslich) to a Greek when the trimeter is broken in themiddle. For the Romans,who had to put up with pitiful substitutes intheir imitation of foreign meters, there would be a caesura, it is true,after KaS81ov;but a Greek hears the verse instead of scanning it. Withthe elision yat' a7ro'XXvuatoAW,he Euripidean verse has no caesura, forthe elision makes the words almost grow together into one. But it is

not necessary for a trimeter to have a caesura, it must merely have nowrong one; that is, for tragedy, it must not be broken in the middle,and must not fall into the three equal parts of which it is made up.Euripides has no such fault (hat keinen solchen fehler); for if anyonecounts in either class Suppl. 303, e. g.aOxLXA yap ev TOltJw( Luov TaX' cvpovwiv,

he understands only scanning. There is a pause after uo'vw,not before.. Aischylos and Sophokles now and then have cacophonous trimeters,only in part as intentional dissonances.Here are several interesting statements. As regards the centralone, that a trimeter bisected by the caesura in the middle soundedhasslich to a Greek, Wilamowitz merely puts the current doctrinein his "hightened and telling way." But the reader of Euripides,however warm his admiration for the poet, may well rub his eyesand wonder if he has read correctly the words, wie er es nie getanhat,-hat keinen solchen fehler. In Wilamowitz's own text of theHerakles occur the following lines:

8 KpE'oVMCVOLKe`w3 7rats,avae ia-Se x0ovo'3.470 es Seetav TC acrjv aAXEe-T?)pLov

$vAov KaOLtctSatSaXov,593 ZX0r7s iuc0Xv 7ro'Xtv 67rEr8'W&rs , opa978 TopvEv/La S-tvov 7roSos, Evavrtov aOaOcts

us a/ukXX?1OOJoyots1256 7rpo's ovETvr7aEts aas avairTvrw re 0o01295 Awvvv yap vast x&ov alrevve7rovaa .Le1301 rTL87T/raCVA 47V 8c; TL KEp8OS 4?fUC.

On none of these has Wilamowitz any metrical note. Puttingaside other lines about which difference of opinion is possible, it



5/23

148 THOMAS D. GOODELLis not easy to see how these seven, in four of which he punctuatesin the precise middle, could all escape his ear throughout all hiswork on his justly admired edition of the play. Is it possiblethat the coryphaei of the "new metric" do not habitually readGreek poetry as verse? Do they content themselves with mark-ing schemes of it, without hearing in imagination the actualsounds of it, which alone constitute in each case the concreterhythm? In some of their melic formulae I confess myself quiteunable, for one, to render or to imagine the rhythmical soundswhich their schemes appear to represent; are they also unable torender or to imagine them'? But every scholar reads the trimeter;it is not clear to me how anyone who is accustomed to readingGreek tragic dialogue as verse can fail to receive the distinctimpression that this particular type of rhythm is by no meansrare. It occurs repeatedly in every extant tragedy, the smallestnumber to a play being seven; the Agamemoni, Oedipus T., andHerakles have each twenty or more.1

Before presenting the evidence for this, some preliminariesshould be cleared up.First, while the heroic hexameter consists of two kola, theiambic trimeter is a single kolon. The poets whose verses have

come down to us were unconscious of any relation, if any suchI The doctoral dissertation of Albert Schmidt, De caesura media in Graecorumtrimetro iambico (Bonn, 1866), is often cited as authoritative. The author does indeedcover the ground with considerable thoroughness, and includes the early and late iam-

bographers, as well as Aristophanes and the tragic and comic fragments. Unfor-tunately his work is vitiated by a fundamental error of method, in that he starts froma petitioprincipii. "Statim patet," he says (p. 6), "quantopere sensus noster, quem inmultis rebus non tam excultum et subtilem quam Graecorum esse constat, offendatur,si trimetrum iainbicum ita recitare cogimur, ut in duas compares et eodem modopronunciatas partes discindatur." The reason, he thinks, is that this violates the firstand essential principle of all art, expressed in Horace's line,denique sit quidvis simplex duimitaxat t unum.

How the unity of the line is destroyed by that caesura more than by others is notevident; nor does he suspect that possibly his confessed inferiority to the Greeks inkeenness of esthetic sense may have led him to a judgment at variance with that ofthe ancients. But obviously the question at issue Is, what the Greek poets thought ofsuch lines. To answer this we must first see how they used them. Having assumedthat the Greeks disliked them, he proceeds to classify the examples for the purpose ofexplaining away all he can and finding excuses for as many as possible of theremainder. Farther, his notion of what the caesura is leads him to include in hisdiscussion many examples which do not belong here, so that my number is smallerthan his.



6/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 149ever existed, between the caesura and any original shorter kola,by the combination of which the longer verse had been at firstconstructed. The caesura has no historical significance, only anartistic one. For the iambic trimeter, then, I use the term"caesura" in the sense of verse-pause only. As in our Englishheroic verse, of Shakspere, Milton, and Tennyson, a more or lessmarked pause in sense within the line is one of several ways ofmodulating the rhythm, so that one simple rhythmic type fur-nishes an endless variety of cadence, and is capable of charmingthe ear throughout the greater part of a play, a tetralogy, threefestal days. By far the commonest place of this pause is thatafter the third arsis; this divides the single verse most pleasingly,when considered apart from verses preceding and following. Thenext commonest place is after the fourth arsis. These two sopreponderated that Hephaistion allows no others: &%oLo'vattO-To,.tat Tov talu,3ov0, Xe'7a C) WEV9f7rEvl/JIpUS&o itOv wo8Cv Ka't00r-/lt/,.pepE Tpt&v jw7ruv.1But even the early iambic poets in theirshort poems needed and used greater freedom than this, placingthe pause occasionally after the second arsis, after the secondthesis, after the first thesis; also many lines have no clear pause-another way of varying the movement. Thus in our best-known poem of Simonides of Amorgos occur the lines:

15 XCXVKfV, Kact /.L8)EV' dj'Opw7rwv opa.50 TV EK yaXvs,)%VTTqvov OL'vpOV y)eVOs.42 opyV/v c bvqv o% rOVTOsaLo2X-vEXEt.86 4tLX- c O'vv OLXCvvrt yV7pac7KEL 7ro0EL,87 TCKOVOra KaXOV KOvVOUacKXvTOV ye'vos88 KapL7rpe7rrs uEV 'v yvvat$L ytyVCTaL.

Farther, a single line may contain two distinct pauses, usuallyone more prominent than the other, sometimes both about equal.In many cases, too, one may doubt; Greek recitation may havediffered from ours in this respect, as it certainly did in others.I should say that lines 86-88 above have no pause, yet onefamiliar with the movement of such lines half makes or imaginesone; we don't know just what the Greeks did. All this appliesto early iambics, in shorter poems. When now the meter wasIP. 148W.



7/23

150 THOMAS D. GOODELLemployed in long poems, as tragedy, still greater variety wasneeded. It was not carelessness, but sound artistic sense, whichled Aischylos and his successors to treat the trimeter withgreater freedom, increasing the frequency of irrational feet andof resolved theses, aild enlarging the range of the pause. Henceeven in the earlier plays we find the pause occurring everywhere,least often after the first and before the last syllable of the line.The following will illustrate the cases not covered in those above:

Prom. 43 aKo9 yap Ov&ev TOV& OpjVdla0atO V Of62 ua O) lOTV) WV 1IOS VWOEOTEpOS.508 cavTOv O aK?)SCL aVc.TVxOVVTOS3 Jsy(910 Opovwv T' alOfToV EK/aXl- 7TraTpOS& dpaPers. 410 vars, Ka7rOOpav'cL 7rcLVTa 4IOLVLOUr?) VEWi486 Kal Awp' a v, M)lX TC KOXTOV, oV

Likewise, precisely as in our blank verse, and for the same rea-son, the treatment at the end of the line varies; usually a pauseoccurs there, more or less distinct, but often the sense is carriedon without a break; a pause near the end or beginning of a linestands in evident relation to such overlapping. To sum up:Judicious modulation of pauses with "the sense variously drawnout from one verse into another" is an elementary artistic lawfor any simple recitative meter employed in a long poem.

Secondly, how does elision affect the question of the caesura?Granted that none of us has heard an ancient Greek recite, sothat vernacular knowledge of what occurred in elision is impos-sible, yet some things are clear. It is no contradiction to saythat in the commonest elisions, as of B, sE, Tad, nd the most fre-quently elided prepositions, the vowel may have completely dis-appeared, precisely as in compounds, while in the less commonelisions a fragment of the "bruised" vowel was heard and felt.Both cases are common in Italian. But no scholar, so far as Iam aware, has maintained that the fragment of vowel soundremaining in elision was considered metrically as a full syllable.Elision was made before such a break in sense as we mark by astrong punctuation, even a period, and even before a change ofspeaker. Precisely how this sounded we do not know; but thereis no reason to suppose that even in such cases the elided vowel



8/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 151was considered metrically a full syllable. There was a sufficientpause, and yet the vowel fragment and the following vowel,separated by the pause, were rhythmically but one syllable. Itis to be remembered that very minute pauses, and delicate varia-tions in the duration of pauses, are made, and their significancefelt, with no conscious effort; also that no one desires, after earlychildhood, a too exact and monotonous observance of the mathe-matical relations of a rhythmic type. The combination of pausewith rhythmic blending (not continuous pronunciation) of thetwo syllables separated by the pause is also common in Italian.Accordingly there is no difference of opinion, on our presentpoint, in regard to such familiar Homeric lines as

A 27 ?) vVv a)O0VVOVT ) V`TTEpOV avO&S toTa,A 37 KXVOl pv, apyVpOTO' '& XpvWnS a&I(nf'`3vKasA 166 xCEpes4Lal &te7rov-', aTrap7/v 7rOTrSaoLo's LKqrTal,E 304 otoLvv3v3poTol dCa' O 8 tLV pE'a 7raXXC Ka oloS.

No one doubts that the caesura, a pause, was felt after the thirdthesis in each of these verses. How is it possible to say that thewords &pyvpo4TO', oS, or Eld' o "almost grow together into one"?Or in the following from the Prometheus:

228 o' 8 olrvEpWTaTL, alTlav KaG VTLVa234 OVK IEV OV8'V', aXA aLTTwYaacoa237 EyW 8 CroXaqO' CE$Av0aa(?v pOTOVS.If these lines contain any caesura at all, it is the penthemimeral;I am not aware that anyone has maintained that elision in suchlines obliterated the break in sense marked by our punctuation.If now such cases are accepted, why should one object to placingthe caesura after the third thesis in such lines as

Prom. 472 7rrovOas aLKiCs r iu' &7r0oo4aXcsi4pcvWvAi. 121 ly}(A jueLvvSev' olS E7ToLKTlpWSC VLVAnt. 74 oTLa7ravovp Ta b-7re'7rwL`v xpo'voHer. 76 &JToVrTlTOV TCKOVT'yw 8( &tacEpuHer. 754 7ra&o-aa8uov yal', a'7roXX/vJaL&6Xw.

That the fragment of vowel remaining in elision made these linesappear slightly different from those in which the same pauseoccurred without elision is probable. If there were any evidence



9/23

152 THOMASD. GOODELLthat such lines without elision were offensive, we might readilygrant that those with elision would be slightly less offensive,because of that fragment of a vowel, though this did not seem tothem a real syllable. But on what ground can we say that suchlines with elision contain no pause'? When successive wordsreally belong closely together in sense and syntax, as article andnoull, noun and adjective, subject and predicate, elision at theclose of the first may bring them still closer together and makethem almost run together into one, as in a compound. But ifthey don't belong together in meaning or syntax, how can theyso run together? So far as I can see, the only ground for denyinga caesura here is the supposed rule that a caesura at this pointis rare and objectionable. If that premise is overthrown bynumerous examples of that caesura, not a few of them in passagesof dignity, pathos, poetic power, careful workmanship, then clearcases of what, on grounds of sense and syntax, would be acceptedas such verse-pauses in the absence of elision, must be acceptedas verse-pauses in spite of elision. Such examples are sufficientlynumerous in all three tragedians.

But, thirdly, it is well to explain what sorts of lines I do notinclude in this category, and why. (a) The mere coincidence ofword-ending with the middle of the line, in the absence of a dis-tinct pause, does not meet my understanding of what a caesura inthe trimeter is. As remarked before, here is room for differenceof opinion as to the presence or absence of pause, and some mightexclude lines which I include. But I exclude on this ground, e. g.Ag. 20 vvv 8' cVTVMS'Yevotr'&iraXXaYV7ro'vwvE. Supp. 699 Kat Cvp7raTaraVTEr /uLeTOV 7raTavTx cTTpaTOVPers. 251 OuZv /4avy?i Kare+9apTaL roXv`Prom. 640 OVK OlO' O7TW VuLV arlcTa7cal LE p,which are accepted by Gleditsch, Rossbach, Masqueray, andWecklein, respectively (see above). So in similar cases. Thereare several like Soph. El. 378 and 892

aX'8E pO)aot 7rTv oiv KaTOL' Eyw.KatL Oq XEaOL 7TaV oUOV KaTEL OuqV.

These I excluded; yet I include (perhaps wrongly) Prom. 625/A27LTOLAE KpS roTVG' 'rep uEXX) 7raElv



10/23

BISECTEDTRIMETERSN ATTIC TRAGEDY 153because the context appears to make a pause after rovi3' moreprobable than in the other lines. In every doubtful case onemust of course endeavor not to be influenced by translation, butto look at the sentence from the Greek side purely; what we areafter is the Greek feeling about it. (b) Lines containing two ormore pauses are not counted, unless the one in the middle isclearly the stronger. Thus in

Prom. 62 'AO- Oo-T?7w 0V LOS VwGTpOSAg. 1394 Xatpov 'av,Et' Xat'po , 6y70 CrvXopthe two pauses appear to be nearly equal; that in the middleseems to me slightly the stronger, but not enough to place it inmy lists. Yet in a few cases, where the first pause of the line isin the middle, the line appears to me to be so clearly divided inhalf thereby that these are included. (c) Finally, comedy andthe fragments have been left out of view, because they complicatethe problem and offer no real assistance toward the settlement ofour main question, as to the frequency and the ethos of thiscaesura in serious verse. Nor have lyric trimeters been included;a considerable fraction of these are sharply divided in the middle.No one doubts that when sung, at least, such trimeters wereentirely conformable to the Greek sense of unity and of beauty.

Here follow, that they may more readily be examined together,all my examples from Aischylos and Sophokles. From each playare given first those without elision at the pause, then those withelision. One or two slight differences are thus made to stand outmore clearly. The text and numbering are those of Sidgwick(Oxford text ed.) and Jebb.

AISCHYLOSSupp. 401 7in7Av&Slundv A&rWAEaasroX v.770 (O&vaTLKTEtWVV$ KV/3pv Tr? aod-o.905 Rteav co&X' ZAdlq abrooa7rdowCas KOS940 rav'ras 8' EKOvOa- /AEv KaT' fCvoav 4PpeV2v234 roS7aro`v 0,uXov ro'v8' &veXXqvo0'roXov711 eU S'AC,qN TpE7/T' dxov'aoat 7rarpos929 dflOVKdXyTOV TOVTr q 6pOV?llLaTL.997 wpav Qxovaas r-qv8' Cort-p7rrov 8porOts.



11/23

154 THOMAS D. GOODELLPers. 319 oYKX7/pasxETOlKOS 7y%, EKEL KaTrE0LTO.465 :Epe-q)g 8' avq pA,)eV KaKU(Wvpv 3aGos489 Kat ?EaOaaXv 7roAXts rEa7ravto-/AEvovs503 axTrvaS ' p/AO, aaao-AEvo0 KVpEL.509 EpjK2V EpctavTES oys 7TOXX^ 7rOV,766 aXXos8' EKELVOVrats TOo Epyov -,vvrYEv-333 &Tasp cfpao-ov Ol ToVT ava0'TpEas 7raXLV403 CXEvPOio'rE 7rarptS', EAEv6EpoTrc 8?493 Xwpav a'4coK cra0', 7r' 'Aetov ro'pov,

607 TOLyap KEXEV0OV T7IqV(8'vEv T'' oXnV/aTo)v784 Ev yap oara4 TO(3' '-', oOl eVV IXKEu;,821 V/3ptLyap ieavOovu' EKap7rzaa-Ev aTaXvv831 X?#at 0Eo8/3Xaa/30vV6' VJ7rEpKo'/LrV Gpaoat.

Sept. 283 EyO) 8Ey' avvopa; EAOLt vv o/A(,,520 a0T7rpyEvotT'av ZEvs E7r'o a1r(os rvX(o)V.1051 &AA'v 7r4oXLtvyEt, a-ivTtua-cv; TapA;426 7rpyOLt 8' aTrLXAE36V, a /. KpaLVOLTVX?7'549 7rVp Ols a7rELtEl oLa' a A q' KpatlVOL OEO'.714 \I0X o\,OV(a-v wr8' '' W/3S(o'as7rvXat;.799 KaMUs EXELi A eXa-r'Ev t 7vXur4aL821 [7rTr oKEv atpa -ya'Vr' AAX wv o'v.]1012 'ErEoKXE'aIA?vTOWv fr' Evota xoovos1058 &XX'v'ro'/3ovXos a-0', arEvvE7r 8' E.

Prom. 976 8o-ot7raOo'vTrEsEVKaKOV(7t UL EKS(tKWg.990 7rpOTpIf'Eat' / E ZEVS yeywvio-at ra8E,472 7ri7rov0a,;tKb? 7?fL' al7rouPaXdEtspEvCJv500 rotaira ufv 8) TaVT' CVEpOE 8EXpovos625 /L0Tt /L6 Kpv7/; Toy6 Oirp AEXXO) 7raOElV.710 7reTa'pa-tot vaoova' br CVKVKots zxotsX810 TOVrov 7rap' Jx6a; Ep4', EUS a\v {$'K

Ag. 339 ToVi Trj a&Xov'acsyEwv 6' 'tSpv'plTa,353 OEovi7rpoEt7rEtv Ev rapao-Kevatouat.509 Vraro's TeX pa ZEv's, o Hv'OLto'savae,567 Tt' Tai'ra rev6EdV6Z; 7rapoiXETat 7rovos640 IrO'XtE Ev EAKo0 EV TO&q/lOV TVXELV,1256 7ra7raE,TovTO 7rp (TVPXETaL 86' AOt.1353 rqSo.ait Tl Spav- TO0?J\ /ALEXtV 8' q4.1396 Tw8' 'av8uata'W v, vrEpStKw /EV o0v.



12/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 155Ag. 833 4nAov rTvOEv'rvXoivr'avEv 400'Vov ao-E/3etv.921 ,-8' paot a-rpwoaao' 'Ert-00ovov 7ropov944 &AA' L8OKELto TavT', ViraL T,t ap/vXas1068 OV.Lq'qVrAXEOctfra- aTPao yao-opat.1221 a-vv EvrE'potSre oa7r Xva , E7rOLKTUYTOV yEO%,1270 Xp'??(TT)pL`av E'T&t7T', brorTevaoas E /AL1302 X 'o6OT X TA'mov'o' a7Tr Ev1r6OAX/ovpevo'q.1360 KatO) TOrOVTOS E7,Tr 8vo-1AfLvXavoi1379 OTr-Ka 8' v0' brato' br' e'6Etpyao-/AEvotL.1420 xtaoaurwv a7rotv; EW7KOOS8 C/OV1446 KEdTat ftXr)TOJP Tp8', E/oL0 8' {r-ya-yvCho. 253 18Zv 7arp0T-t cot, 7rarpoarEpv,yovov,699 iarpoS ZXrtLs')v, 7rpo8ov(oavEyypa4E.909 7raTpoKrovovo-a yap O-VVOLKrq(TElg EMOL;9 ov' CeTEtva XEp' E'' E'Koopa VEKpOV130 XEAywaXovoa 7raTEp EITOLKTE pOV0 9.L243 7no-rs 8' a&EX4os o6', E'/LoL Eas 4,Epov.501 tLoOVVEOMfOVq rov'8' f4.E'vovq r6ax.

545 Kat /aX-Trov aE/tx'aK .LAOVOpE7TTr7ptOV,561 vzw vv avopt rT8w E4' EpKEtOVq 7nVAs752 Ka' 7roAAa KatLAOX0 p av-a EAr'T EAot919 ,u? 'EVyXE TOV 7roVOVVTr'E(T( KaO07l.Lv7u).Eum. 87 a-Ovos8e 7roLtLv EVl 4EpEyyVOv TO ao'v.116 ovap yap v4tLas vvV KXvraLMATrpaKaXW.282 7roTatvtov yap Ov 7rpoS co-TLa 0GEov

618 0 /A77 KEXEv'oat Evi 'OXvpurCwvraTr-p.829 A' ovi8evv'ToV 8dEL a 8' EVXLS E/OL848 opYas evVOLta(L aot0L yEpaLTEpa yap Et.591 ELrELV yE /t&EVTOL 8E a oJW13 KaTEKTaves.643 u/aw 8' oCovEVTavrT E-yw pTavpopat.685 7rayov 8' 'Apetov TOy8' 'APato'vwvSpav901 Totyap KaTra XOOv'OvaT' E7rTLKqrYCE 4t"Xovg.906 cEvAsVWrVEOVTT'7rLtoT-LXCLVXOova

If some of the above appear doubtful, it may be noted that, inaddition to those mentioned earlier (pp. 156, 157), the followinghave been rejected as falling under class (a) or (b) (p. 156): Supp.300, Sept. 385, Prom. 787, Ag. 258, 955, 1435, Cho. 489, 522,



13/23

156 THOMASD. GOODELLEum. 619. I merely wish to make it clear that no desire toswell the lists has consciously influenced me. There was no needof that. SOPHOKLES

Ai. 86 -yevotro uEvrav .7av Amo rEXv@(ouVOV.9i5 E/3a/as EsyXogEv 7rpos 'ApyJ wv -rparTw;343 XEqXar'o-EPO'VOV;EY(O O a7roxXvPat.437 Ey7 O' 0 KfVOV 7ra's, TOVavrov ES TO7rOV651 83acM} o-' pos Us, 'LqXv'vOqv o-rO1125 e'vv TrL &tKaiw yap uLEy'kEc-rtv 4povetv.1129 ,U vvv I O9sov',OEoZs o)ovos.

12 52 A' oiS)povovvreg EV KparovaL ravraXov.1253 puyas S\ 7rXEvpa38ovsvro\ tKpas o6uOs1377 ocrov r&Or'XGp\s , roco'v' etvat 4XAos.

121 y\o u\v ov'oev ot' bErotKrtpw SE VtV281 3 1o 'XOVT(I)V T-)VO& t`n'Tqraqo0a` CPG292 o 8' ve 7rpos ,UcPat',act 8' vuvovucva294 KcIy\j ua0oGV& E7 6 'Sy va' uo'voS.513 KELVOL)TE Ka/uOt Toy6' 0 rav 06 ,vqiVt.768 KpaTOS KaTaKT?7)atT ey\ oc KaM StXa780 rota^36 'o pJVr" 4T ' O 0 evavS E' opas785 opa ,uoXovaa TrovSo&n-o7o OpoE?.810 dA Eut Kay'/ KELT' 07rOt7rEp av (OCEV(W.919 vrX)y7/r6,a(vX\v at4l' a' OLKE'as o4ayn^s.

Lines 574 and 1385 are omitted; also 690, because, in spite ofverbal similarity to 810, it seems likely an actor would delay atrifle after EdFL'and little or none after ECKdE''.

Ant. 55 rpt&ov8' aEX4)w\ ovo ulaV Ka' -pEpav518 7ropOwvS\ r WvoEyTv oo' aVrrtras v7rEp.555 (TV /uEVyap ETXOv6V, Ey\ E\KarOaVetv.723 Kal TWv X,EovrWV Ev)KaXO\vTO\Ma=v0dvtv.44 7 yap voZts Oa'TETv -o', a7ro'pp7Tov 7roAXE;57 KOtVOVKaTetpyaffavT e7raXXotv XEpoltv.74 oota ravovpyqrcac' crct 7rXEt'WvXPovos77 ra rTov 064V EvrT arTuaao' EXE.80 aTv\ uc\v TaO' av rTpOVXO't) y\ T\ ra4ov399 Ka\ KpLVEKO4EXEYX' $Y\' OXEvOEpo407 rotoVrov 7v rTOrpay/A' * T0W yap 7KOLEV764 roVi\v 7rpoxT I/EL KpaT EVo4OaFA6ous p(v,



14/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 157El. 67 aAX', 7rarpwoa , 0Eot T' 7Yxwptot,1036 arqtuag pEv ov, rpo,uqvt)ag S o-ovi.1038 Jrav y'ap Ev povfrs, T'O0yCt oV V@^v.

1205 u'OEs OO'ayyog vviv,Jrw, TOrrav /LaGfls.44 Xo"yw Xpp' rou& JTt $EVag pv ct59 Tt yap /LEXvvrELtroi3', Jrav Xo'yw,Oav(v360 ,uCXXotts ot'q-v Sw^p', C4' otot VvVx ka's,

678 ovi uciv Ta aavnrp 7rpa^o-', CLOL TV',vE'VE696 Kat ravra uf`v TOtaVGO Jrav SC Tts 0Eov923 7rw o' OVK (7) KaTOU' a' y/' itOV c9L4avwU;938 ov'TrwS c'Xct (7-t TavT * av oSepot rtp,o,1044 .XX' E 7rorTTSTavT cravEOCts Ci& C.1056 cApovEZv,fpovct Totav' * Jrav y"ap CV KaXOtS

1124 cvSvcr/evet'a y' ov'c' c'rareitat TarcX,1302 KaL TOVtLOV cTTaL T?78' re Ta's oov"as1338 (v TO6s OtOVTOLS JT', arAXa'xat S 'o a,K.1480 'o OVK OpE0T77s E'TO' 6rpoo vwwv 4c'.

0. T. 12 E,UoV rpOOapKcLv 7rav ocEXyvrov yap av110 CV TO ECa(TKC79. TO 8( tToV/eVOV130 v otXOe8o's>Y,L TOrpos wTo0TTKo0e-V141 KfVt'v 7poV0apKWV OVV0/aVTov w EX.968 KEV9IL KaTW Syn) S y 8'& 8' 4aSaE979 e'K?) KpaTL(ToTV 6vV, O0Trw UvaoTo TtL.

1066 KaL /.L`v Apovovra y' ev Tar X4o-rca ro&t cyw.1155 SvCr?Vo0, avrt Tov; Tt 7rpooXPJ),wv /uaCetv;

46 L&',( f3pOTWV aptoT, (avop9O(ov 7roXtv'250 cv rots e/lAos y(voLT' e'/ov avV(L&)Tos%328 raLvrEg yap o0 (fpov^LT'E7 o o01) IrOTE

364 JrwrL t )ra KaXX', LV' Op7t 7ov419 J3sXrovTavvv /Ev op6O', e7rTa S o-K548 rToT' avro /L7 (Lot pag', orws OVK EL KOS.779 a&r"p yap 'v 8et7rvots (' v7rpVArX710TGE /L6E%786 EKVtgE a/2Et roVO' VZ0lpre yap 7roXV'.1058 OVK av (VvoLTo rovT', OVrW.s Iy0 Xaf3zv1061 K MEL, p,a, ,s Ey.

1156 Tov 7TaL8' I'E&Kag Srw^8 ov ovTOg tcfTOpEL;1163 MCIOViEV OVK6y7y', fe&&q.v/ E TOV.1442 ovTrg EAXEXO)avO' oJlvs o', tv' TTaFev1458 aAA 1 EV 7)/AV /Otp', OTOLTEp Etq', ZTW.



15/23

158 THOMAS D. GOODELLTr. 163 /LoLpav 7iarpoag yv? 8tatpErov V4lOL,566 x( ZyvvoEvOvS 7rats ,rtTpi`as XEPOTV761 XctasarapX' v 8ovi3 aTap TEa raaVO' o'/ov-45 ur?vas 7rpos E oXX0 7rEVT aK 7 VKTOS EVEL.178 Ev(frq/utEav VVV L'o-X &ETE KaTao-Tre472 aX, (AtAnrbsroov', -E7 uE avoaEvUo614 KaEtTrOW8' al7rot0(Es (Tq,u', 0 KETVWOV/La0ES667 KaKOV uEY' EK7rpEL$ao' abr' DXrta'8o0aX,s.737 TrwV VVV 7rapovoTWv TrWvS' a',uE/E/afOat 7TroOEv.917 7rTwS' ETErEETOVT, E7TEV0p0VET'Vo

927 Kayl SpOLt'a 83aT', o0Tov7rEp EETGEVOV,1098 Atov TpLKpaVoV oTKVaaK', arpoouaXov TEpas1100 SpaKovTra u-VX(ov fVAaK' E7rT o(xXarot0 To7rOtl.Phil. 15 aX' Epyo0Vor oTOvraL O)' v7r-7pErELv,57 XE'YEtv, 'AXLXXE'WLLs TO8 OVXtKErT'EOV121 q 1Lv?)(0ovEvEtLvv a oot 7rap7}vvEca;263 o TOV Ilot'avrog waLs LCXoKTrqrT,, 0'v VL297 iErAv'AVTa Vov O KaEt mTaE /2, aEt.

366 Xos KparTVIvELiv, O AaE'prov yolvos.389 Xo'yosEaXEKTr wat 8' 'ArpEtSag oTTV7OV420 XAXovTrE`LEt viv &V ApyEcWv (TTpaTp435 XOVTOSrEOV?)KOS lV Xoyw SCo' iv 83paXEL589 'E,u. opaE TL7rotLs, 7raL. NE. CTKO7rWayw) 7rELXat.907 OVKOVVV ots yE opa1SL7 Ev oS 8'LvJ'aJs, OKVW.1009 avat0ov u'V o0, KarTELcov 8' f(uoOv1021. oJvuEpv'yv/,Oag ;v, Syo 8' AXyv'vo,ua1040 aXX',(0W 7raEpL(oa 0EOELT' (47rof/tO,1049 o0v yap ToL0VTWV 8EL, TOOVTOS EEL' EY7.1064 0WAOLEtLOOT/L7/0fEL v 'Apyetots ckavEZ;1237 TL (S, 'AXtXXE'(sWaLL; TLv E'tp7Kag Xo'yov;1261 or 8', ll'WElotavWros aLt tLoKrTV)TvXEyw,1274 TrorEpa 8KowratE (Tot uJEvovrTLKapTepetv,1298 EaVv 'AX"AtXXEw;ra, 'Eav rE upW1396 g#~v, JJWEp8?W 7 qg, avev (owrptag.

226 SEt1ETVrET EKWrayV/T' arW7ypta),uEvov238 y7EyvE /uot Wav rovO', o0rwg clo0 TrLsEr.342 av'OtsWrEAv(Lot Wrpacy-/', O'TW ar' EVV/3ptLav.346 s0o0ov)OE/Jtg yt7yVOLT, TEt KELT(00TO359 KLV0S /LE/VOVV KETr' (70) 8' O 8V OpO,477 -oo8'OKX7rOVTrOTr', OVEL013OV KaAXOV,



16/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 159Phil. 570 KElVO's Dn' AAov vop' I Tv83 s rE7a-613 ayOLVTO VVqOOV T-q(3' Ef' V7s ate ra vvv.617 OLOLTO FLEV /aXL(T6' EKOVtLOVXafl`v,630 (3Z$amIE\s dyovr' 'v 'ApydoL's-UOLS;641 aELKaXo\ 7rXoOsaO', orav 0,Ev'yy KW(a.666 EX0p0vIL' JVEpOEV Jvr' ave&ro--as 7repa.957 Oav\v 7rape4 8aZO'vf' uv f Ap/73oyrv,990 ZEV%,) 8OEOKTal TavO' V'7r-pETOJ (3' '[email protected] a7ELV '7r' W(Tr T7)r0o EV 7 LE 7rpoV/3c8Aov1037 ZoLSa o' d AEXEL y'* E7rEL OV7raT'v oi-o'Xov1056 Ta y' 507 EXovrEg avT' *rE\ 7rapE(TTL 1LEV1437 OVTOSE\ KaL av TOV' *'y\ (3' 'Ao-Kr7plLO\V0. K. 480 7r0VTOVSE 7rX)o-as KA; L88ao-KEaL\ TO&S.

568 OV(V 7rX0ov FLOLTOv METE(TTLV i/LLEpas.607 1 cLfrTaT' ALytEWs 7raLt,OVOLSV L7tvETaL610 OLVIELJAEV aXV3 YyS, /OL'VIELOK Tw`LaTOs,1038 xJpv aVrdiEL v v 3' ZLv,OiL7rov3,1435 oa* 6' EVOSOL1 Zv's, Ta(3' EL OaVO'VTL /LOL1489 av0 wV 7rao-xov Ev', TeXEo0pov XaPLV52 Tts &a X6 pos 83?T' (V q' 83E/8rKa/JEv;266 oL'OV3 Ta'pya Ta(' 0rEL Ta y' cpya yov288 OV-qcTLV-TTO^L TOUIT(3'T av 8' o KVpLOS462 aZro6' rE 7raZt3Es 6' aLoS' ErEL SE\ TrflT(3EyVS575 TOV1T'aVro vVV SL&aTx', O7ran av EKJaL0.1125 aiZri TE Kal y7 T?(O' (7rE TO ' EVeftE\1171 E$0oL' axovwv T(t)VO' 'OS ( 7rpooTaT7rq.1275 @ o-7rrpp/aawr' avopO\s rov(3, 4.aL 8'O6/a"LOVES,1407 IL7) TOL IE 7rpOs 06EV (T4Uw ', av al TOVS' a'pa\1429 ovo' ayycXovL77 ,Ev faVp' E7TEL oT-paTrXaTov1542 Awrales, o'E(re7O' T (y0) 7ya\p77yEIw\v

For Euripides the continuity of practice with his predecessors,and his one marked peculiarity, will be sufficiently brought outby complete lists from four plays with complete references to therest. The Andromache, Medea, Bacchae, and Iphigenia at Auliswill represent his earlier and his latest years. The numbering isthat of Nauck (Teubner text) and of Prinz-Wecklein.

Andr. 47 0os 0' E(TT 7ra OL MOVOS% V7'rE7ur7rw A'Opa247 FLUforVV 7E 7rarpL&a \'v 'AXtXXE'Ws'v



17/23

160 THOMAS D. GOODELLAndr. 656 EKrOPOSa'dEX4s v, Sapap 8' r8' 'EKropoS.698 ov''ev 7rXcovSpwvEvosEXEL7rXEt Xo'yov.

969 yvva^tx' o(Foa SoVo Vi7r&xcO'OIEpov973 yauovg dcOevat aovs, EsaXAEywvTvXag

1090 'Aya4Evovog 8e 7ra^tgm-TEL'Xwv 7roXLv1117 Xu v KarT'OHL o-raTs7rpo-e vXcrat Oe4;1268 EXOOJo-LTy'v a-ov TO yap 7rE7rpwpEVOv17 ov'yXopra vati 7OL', l'V' / OaXaow-L'a41 KaL VVV KaT' OLKOVSTT', abro :rarprv uoXw"V

216 e'vpavvov coxXs avp'V' 'V v yEpEL WX0o231 e yELvTpO4rovs pq TOMv, tooaLg &vlTL VoVSi.309 -K() Xa/3wv o-Ov raLS', 8vets !Xovg Solovs373 avopos' ~~awpTaVOV- a,uapTavuL ftov.390 7rotov8' brp/o-a &i4&'; EKOLfL AqV flta402 KO/SL7/- 7rLT7rao06ELO' ETreL8' a'cJ,uo'y,v433 X' Ep7r' es OlKOVS TOV'c8, V E's e Ev6vEpov441 v Ka' v0o0o0v TO'V8', V7rO 7eTEpOVorauas;868 oVT' avi To VVV aov SZeM'0o eLptmU,Vl ayav886 /aVTrck Aw8wVaV'- bre 8' acOLKo'FA?,v890 vaLova a'4' r v 7riSL' o',wss o-Artv4LA7q.966 ri4wv o- abr OLKcV TiVS'. L) y'ap ovTa rptv994 JU-8V 0,o/370VTraeZ8', Joa CL" *4i' vfpLUCv.1145 IrCTpaLOLv avTKXAy' - EV EV8La 8 7rwS

Med. 266 OVx'KOTLVXX-1 .p7)v FuaLSoverE!pa.701 88WO-L8'aVTr4 TLS; 7repaLVE MOLXo'yov.729 a -So8e' avT7' &raAXao-ov roSa.19 yta; Kpeovroq 7rai', 0S atavuva XOovo..q o]3&c LV EO'T~W~380 0Ly)p8oFoV,tuZBaZ apUa f470 e1nAMovsaXK(2) 8pavavT' evavTLoV f,tX7ll,

(Wecklein's conjecture is quite needless.)551 (Irei /LeCTOTV/veVp' 'IwcKLcaXOovO's618 KaKOV yap avspos 8owp' OV/O-LV OVK EXEL.669 7rai8ov ipevV(^v TrEpL' J7rAS 7EVOLTO IOL.805 VVLOg TeKv(eo-e 7rat8', brt'i KCU7-V KaKWJ3947 7r' yap avrj 8xp' a KaXXLOTEveTaL1014 Kay(e KaK(Ow 4pOVOvO' E,qXavq,oa4Lnv'

1060 oVToLrOTrJcrTal TOV'O' 07rWS!X6pOp i(1141 KVVEL8' O I/IEV TLS XEZp, ? 8N $avOOv Kapa

Bacch. 49 TaVOV8Ee 09EVOSlE FLETrvoTIW 7r0Sa,



18/23

BISECTEDTRIMETERSN ATTIC TRAGEDY 161Bacch. 265 'EXt'ovos 8' 'v ra.ts KaTatcrxvvats ye'os;298 IaVTLs 8 O &LMa Wv O O yap'y"Wt'OV353 TOv09XVAopc/ov VVOV, Og ELafpEL vo-ov682 ui-p 'Aya,n oy, Tp&Tov8' 'Ivuoopov.841 `8ovisCp',ovs 7pS v yx 8' Xyyoao".922 &AA' 7rOT i7cr0a O-4p; TeTavpxaat yap ovv.975 TOv8' c1S a&yw^valdyav, o vuoKRawV 8' Cyxo

251 vap0?7KLPaKxXCoi'T' a'o,vV0at, 7raTEp,254 OVp(JOVLO7cTELs XIP" If'/VPS TpOS 7TTEp;347 DAXOve 0acovs TOV8', LV' OLWVOOTKO7TEt448 KXA78EsaIvrKav peTrp' IV v OY7Tv-rs XcpOS.451 GtE(0E XEtpw^v roV8' EvapKVTLV yaV Wv696 ve/3ptSas T' aVET-rdiavO' ocTaLcTLVQ4,NaTrv704 OVfXTOV8Cts Xa/3ovr' b7raLUEV ES 7rCTpaV,1261 (v T(8' at /EVELT' 'V p KaOEOTaTE,

I. A. 468 yr7Las o Ilptalelov Hapug, os ELpyaara Tra8e.668 CT CTTl Kat 0oL7TXovs, 'va iv?a,l? varpos.733 e/I 7rapew 4s o vvOlots 7rp`7ft.747 KOltv TOTiS OeOV PtAov, LOl 8' OVK Vxeg.827 A'8as ILEVe(l/t 7ral% KYVATra17Y(rTpa 8E' MOL1461 KX.7r AwXvEXo i4'vi1 Amo', Tp, rLOOV,

1593 7rpOvqKIE83w/Ut'av,IEXcaOVopEL8pO,uov;309 aJXXots /LLXXO TaVT' -4aEs &e T-7V8' IoL.930 aAX' cv0'8' ev Tpot'a T' XEv09Epav4Otv1130 ov'&Ev EXAEVL,UOVET,' * Ep(i)Ta^orOa&G1EA.

1153f Kal TW 4 lOS Te 7TaL8' E/A(A) TE O`UyyOVwLr7rOt:Tt /lappaLpOVT' ELrTOpTEaTCVaCTVv1463 ApT'u8o' EL XEL/LOV', 0TO1 (O4)ayoqCrO,LUL.

1579 Xatu/OV T' E7rT(rKo7reZO, tLva WrXyteEv Qv'Eveni if one insist on striking out a few lines from the abovelists, enough will still remain to make it impossible for a candidobserver to say that such lines are rare, or were really avoided bythe Greek poets whom we most admire.Examination of the examples brings out clearly three points,of which the second and third were noted by A. Schmidt.First, those without elision at the pause are rather lessnumerous than those with elision; but not enough to lend anyreal support to the idea that the former were less agreeable. In



19/23

162 THOMASD. GOODELLsome plays (Sept., Prom., Cho., S. El., Trach., lIed.) the differ-ence is really great. In others the numbers are equal or nearlyso; in two (Eum., Phil.) those without elision slightly prepon-derate. No one could maintain that those plays in which theratio of lines without elision is smallest are in general better ormore carefully written plays than the Persians, Eumenides, Aias,Philokletes, Bacchae.

Secondly, one notes immediately that in many lines the wordbefore the pause is a monosyllable, which brings a word-enldingin the place of the most frequent pause. There is probably somesignificance in this; the poets liked the flow of the line better so.Plailnly, also, this preference increased. Moreover, there is a dis-tinct difference in this respect between those without and thosewith elision; the ratio of monosyllables constituting the thirdthesis is greater in the former. Just why elision after the thirdthesis should lessen the desire for word-ending before that thesisis not easy to see. It certainly looks at first as if some hint ofeither a penthemimeral or a hephthemimeral pause was desired inlines broken in the middle. The word-ending hinted at theformer, the elided syllable hinted at the latter, though in neithercase was the suggestion more than hint. Apparently, then, if thehephthemimeral pause was thus vaguely suggested, there was lessneed of hinting at the other. And it appears that Aischylos inhis earliest extant plays cared little for these pseudo-caesuras, inhis later ones much more. His later practice gave the note to hissuccessors, Euripides following it the most strictly. But onfarther examination there rises a vigorous doubt whether thisliking for a monosyllable in the third thesis has much coilnectionwith the caesura media. For if one will watch carefully through afew hundred trimeters for the instances of a monosyllable in thatplace, one will be surprised at their number. They are manytimes more numerous than the lines broken in the middle. Inthe Agamemnon, for example, even omitting all cases where themonosyllable in question is a proclitic, or a preposition followedby its noun, or a form of the article, or a word followed by anenclitic (unless the enclitic, by elision, ceases to be metrically asyllable), we find that more than one trimeter in seven has a



20/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 163monosyllable filling the third thesis. Only thirteen of thesemonosyllables, less than a tenth of the whole number, are fol-lowed by the main caesura. Clearly the relation between thebisecting verse-pause and the monosyllabic third thesis is not sosimple and direct as has been supposed. These are among thesubtleties of Greek verse-construction which our foreign ears havenot yet mastered.

Another allied phenomenon is the disyllabic third thesis, apyrrhic word just before the caesura. This appears first in Pers.403, then in Cho. 130; these are all the Aischylean cases includedin my lists. Sophokles has it in Ai. 343, Ant. 55, Tr. 1098 and1100-two lines apart and rime-words. Euripides made it oneof his metrical mannerisms; the two older plays here representedcontain three cases, the two late ones contain six. For theremaining the numbers are: Alk. Heck., H'kleid., Rhes., none;Hipp., 1; Supp., 2; Troad., 2; I. T., 4; Her., 5; El., 6; Ion,6; Or., 7; Phoen., 7; Hel., 9. That these figures have somerelation to dates, though the relation is niot to be pressed toohard, is evident; also that the phenomelnon is a special form ofthe broader fact that the third thesis in general is oftener resolvedthan any other, even in Aischylos.

That, however, the relative frequency of bisected trimetersbears no clear relation to date will be evident from the followingtable.' No weight should be laid on the precise figures, because

' That the reader may test the table if he cares to, the remaining references toEuripides are here given. Those before the dash do not have elision at the caesura,those after the dash do have elision; monosyllabic third thesis is indicated by a, disyl-labic third thesis by b.Alk. 287a, 789a, 939a,-1, 8a, 71a, 179, 289a, 379a, 522a, 625, 633a, 809a, 955a.H'kleid. 238a, 424a, 837a,- 62a, 66a, 161a, 5l1a, 516a, 551a, 650a, 663a, 725a, 824a,1019a, 1047a.Hipp. lOa, 888a, 902a, 965a, 1163a, 1319a, 1426a,-322a, 356a, 604a, 1042a, 1181a,1190a, 1457a.Hek. 15a! 37a, 232a, 253a, 265a, 321a, 879a, 979a, 1133a,- 254a, 301a, 387, 537a,964a, 1125a, 1169a.Supp. 231a, 268a, 511a, 653b,704a, 754a, 1060b,1088a,-8, 195, 466a, 520a, 741, 843a,858a, 1098a.Her. (see above, p. 151),-2, 76, 153a, 174, 456, 537a, 631a, 754a, 931a, 984a, 1124a,1152,1221,1402a.Ion. 646a, 742b,999a, 1019a, 1028b,1030b,1041a, 1342b,1370a, 1527,- 16a, 39a, 315a,372a, 576b, 620a, 633b. 847a, 851a, 933, 1178, 1205, 1295a, 1333a, 1371.

(Footnote contintted on next page.)



21/23

164 THOMASD. GOODELLthe inclusion or omission of a line or two in a play alters ratiosmaterially; but if one endeavors in making the lists to be guidedby the same principles throughout, the subjective element will notgreatly disturb the comparison, if that is not pressed in details.Fractions are disregarded.

RATIO OF BISECTED TRIMETERS TO ALL TRIMETERSAISCHYLOS SOPHOKLES EURIPIDESSupp. 1-60 Aias 1-50 Alk. 1-58 Tro. 1-66Pers. 1-33 Aut. 1-77 AMed. 1-74 El. 1-51Sept. 1-55 El. 1-66 H'kleid. 1-59 L T. 1-85Prom. 1-92 0. T. 1-52 Hipp. 1-71 Hel. 1-45Again. 1-50 Trach. 1-74 Andr. 1-46 Phoen. 1-48Cho. 1-56 Phil. 1-27 Hek. 1-59 Or. 1-48Eum. 1-58 0. K. 1-70 Supp. 1-57 Bacch. 1-58Her. 1-38 I. A. 1-57Ion 1-42 Rhes. 1-84

In the Persians, an early play, Aischylos made a larger pro-portion of such trimeters than appears in any other except thePhiloktetes; and beside the latter in date stands the Oediputs atKolonos with but one in seventy. The Promnetheus has thesmallest proportion; the difference between the Aischylean Sup-pliants and Eumenides is imperceptible, and with these go theBacchae and Iphigenia at Aulis. Curiously, the Herakles turnsout to be precisely the one play of Euripides in which he includedthe largest proportion of such lines. And what becomes of Euri-pides' yr6ssere Soryfalt? If the lines with elision were disre-garded, these relations would not be essentially changed; allthree tragedians employed the type with about equal frequency.Finally, we return for a moment to the really central questionof ethos. No single sharply defined significance can be attributedto this or any other formula of the language of rhythm, though it

Tro. 372a, 386a, 619a, 650b, 988a, 1177b, 1275a,- 9, 658, 922, 1013, 1285.El. 43b, 248a, 840a, 1042a, 1094b, 1273b, 1275a,-14b, 78a, 96a, 305, 332, 416a, 837b,979a, 980b, 1012, 1087, 1262a.1. T. 87a, 370b, 484b, 496b, 674a, 1014a, 1040b,-27, 547a, 664, 1002, 1036, 1051a.Hel. 86, 290b, 412b, 449b, 503a, 575a, 585a, 605a, 987a, 1028b, 1236b, 1241b, 1399b,

1410a, 1449b, 1520b,- 24, 49a, 102a, 116a, 938, 984a, 1219, 1237a, 1438a, 1546a, 1574, 1610.Phoen. 46b, 449b, 521a, 574a, 738a, 760a, 761a, 836b, 846b, 928a, 1090a, 1091a, 1317a,1349a, 1400b, 1649a,-19a, 64b, 74, 571a, 768a, 922, 1006a, 1223, 1608b.Or. 35b, 63b, 401b, 549b, 624b, 1053a, 1076b, 1220a, 1585b,- 12, 230a, 489a, 533a, 559a,879a, 1032, 1040a, 1054a, 1079, 1169a, 1189a, 1239. 1328a. 1342a.Rhe. 161a, 388a, 579a,-190a, 664a, 857a, 868, 965a.



22/23

BISECTED TRIMETERS IN ATTIC TRAGEDY 165is true, as Headlam has shown, that special formulae in lyric werehighly suggestive of particular themes. But, negatively first, itis obvious that a formula so freely used by all the tragedians cannot be set aside as faulty or as hcisslich. Is there not a touch ofthe comic in the spectacle of modern lVissenschqft lecturing asupreme artist like Sophokles on his proneness to bad meter ? Atthe least this may be affirmed, in general form. In verse, in alllanguages, some things that would be bad if used to excess arenot bad at all, but contribute to the beauty of the whole, and aretherefore good and a mark of skilful workmanship, when employedin proper places and in due proportioni. The bisecting pause inthe trimeter is one of these things in Greek tragic dialogue. Wemay put with it two other things that are severely condemned bycritics, which are nevertheless approved by the poets. I meanthe division of the trimeter, by pauses, into its constituentdipodies, and the absence of word-ending within the foot through-out the line. Examiples of these are:

S. El. 282 ey` 8' pGr v &va/lopos KaTa rTeyagPhil. 807 aAX', W TKvov, KaL Oapfos L7x('* U rsqe /Uot0. K. 1169 CL) ()XTrare, (7XEs v7rep et. @-q. TL S (C't got;Aisch. Supp. 401 e7rT?7Xv8asTL/yV a7rw'Xeaa 7roXLv.Pers. 469 7reT(3 7rapayyetXas alap acrpaT(v/JaTLProm. 612 7rvpos /pOTOLs Sorqp' opa's llpo/lA)0Ea.Ag. 955 aJvos, OTpaTov &Up7/u', EuOL $vyf'0a7rEfO.S. Ant. 57 KOLVOYaT(Lp-yaoavT' i7raXX?OLv XEpoTY.

See also in the above lists Aisch. Supp. 905, Pers. 465, 509, 831,Ag.1420,Eum. 906,Eur. Andr. 373, 402, Med. 470, 1014.As occasional means of varying the movement of a single

rhythmic type continued through a long poem, these are allacceptable. Beyond this it is perhaps not easy to prove any-thing for the bisected trimeter.Yet I think we can go farther. Such trimeters are employedin a great variety of circumstances and moods-serious, pathetic,prayerful, as well as argumentative or quarrelsome. Thus by thisdivision parallels or strong contrasts are emphasized in Aisch.Supp. 401; Prom. 976; Sept. 1051; Aias 1129, 1252 and 1253(two cases in successive lines), 1377; Ant. 55, 77, 80, 518, 555;



23/23

166 THOMASD. GOODELLS. El. 1036; 0. T. 141, 419, 968; Phil. 907, 1009, 1021, 1298;0. K. 610, 1489; and often in Euripides. In prayer or entreatywe find it in Sept. 426, 549, 714; Prome. 625; Ag. 509, 921;C/to. 130; Etrn. 906; S. El. 67, 1205; 0. T. 46; Phil. 1040;0. K. 1435; Ion 576, 1333; I. T. 547. We find such trimetersas closing lines of fine niEt9. Several passages of unusual eleva-tion of tone contain one or more. The solemn opening speech ofKinig Oedipus, royal in dignity and compassion, ends with thewords:

e,LOV IrpOOcapKEtV 7rav- SV0aXylqTOS yap aveGlV TOtaV8E /Lq' ov1 KaotxKTtpwv e8pav.In the pathetic appeal of the priest which follows is the lineL0', c /3poTWv acptoT', cvop0woov roXtv.

Ismene's entreaty to her sister not to throw away her life in thevain endeavor to bury her brother contains two such verses;Antigone closes her answer to this entreaty with the lines:oOta 7ravovpyqo-ao-' bre't 7rXELWVXpOVOSOv et ,L' apeCTKeLVsL' KaTw T-WVvOa8e.EKELyap a'ct KE0/lat - 00 S' El 8OKE,Ta Twv OcEV EVTL/A'TLtLaaas EXE.

Hipp. 1436 is part of the pathetic consolatory prophecy of Arte-mis. Pers. 403, with disyllabic third thesis, is part of theinspiring battle-cry of the Greeks at Salamis; Eum. 848 is thesoft answer of Athena to soothe the anger of the furies. In theseand like passages a great poet could use the equal balance of lineto produce a peculiarly musical effect, enhancing thereby thedignity, pathos, and beauty of poetic expression.The English schoolboy is properly taught to avoid this caesurain his iambics, as the least frequent of all. But the scholarwho has gone through the English training in writing iambicsapproaches tragic dialogue with better understanding of thepoet's point of view. He has acquired an ear for the Greekcadences; accordingly he does not hesitate to use this caesuratoo upon occasion. The admirable translations in the NovaAnthologia Oxoniensis contain several tasteful examples of it.

YALE UNIVERSITY

bisected trimeters in attic tragedy

Documents